Repositioning Antitubercular 6-Nitro-2,3 ... - ACS Publications

J. Med. Chem. , 2016, 59 (6), pp 2530–2550 ... Publication Date (Web): February 22, 2016. Copyright © 2016 .... Journal of Fluorine Chemistry 2016 ...
0 downloads 0 Views 806KB Size
Subscriber access provided by GAZI UNIV

Article

Repositioning antitubercular 6-nitro-2,3-dihydroimidazo[2,1-b] [1,3]oxazoles for neglected tropical diseases: structure-activity studies on a preclinical candidate for visceral leishmaniasis Andrew M. Thompson, Patrick D. O'Connor, Adrian Blaser, Vanessa Yardley, Louis Maes, Suman Gupta, Delphine Launay, Denis Martin, Scott G. Franzblau, Baojie Wan, Yuehong Wang, Zhenkun Ma, and William A. Denny J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01699 • Publication Date (Web): 22 Feb 2016 Downloaded from http://pubs.acs.org on February 23, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Medicinal Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Repositioning antitubercular 6-nitro-2,3dihydroimidazo[2,1-b][1,3]oxazoles for neglected tropical diseases: structure-activity studies on a preclinical candidate for visceral leishmaniasis Andrew M. Thompson,1,* Patrick D. O’Connor,1 Adrian Blaser,1 Vanessa Yardley,2 Louis Maes,3 Suman Gupta,4 Delphine Launay,5 Denis Martin,5 Scott G. Franzblau,6 Baojie Wan,6 Yuehong Wang,6 Zhenkun Ma,7 and William A. Denny1

1

Auckland Cancer Society Research Centre, School of Medical Sciences, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand; 2Faculty of Infectious &

Tropical Diseases, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, United Kingdom; 3Laboratory for Microbiology, Parasitology and Hygiene, Faculty of Pharmaceutical, Biomedical and Veterinary Sciences, University of Antwerp, Universiteitsplein 1, B-2610 Antwerp, Belgium; 4Division of Parasitology, CSIR-Central Drug Research Institute, Lucknow 226031, India; 5Drugs for Neglected Diseases initiative, 15 Chemin Louis Dunant, 1202 Geneva, Switzerland; 6Institute for Tuberculosis Research, College of Pharmacy, University of Illinois at Chicago, 833 South Wood Street, Chicago, Illinois 60612, USA; 7Global Alliance for TB Drug Development, 40 Wall St, New York 10005, USA

1 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 72

Abstract

6-Nitro-2,3-dihydroimidazo[2,1-b][1,3]oxazole

derivatives

were

initially

studied

for

tuberculosis within a backup program for the clinical trial agent pretomanid (PA-824). Phenotypic screening of representative examples against kinetoplastid diseases unexpectedly led to the identification of DNDI-VL-2098 as a potential first-in-class drug candidate for visceral leishmaniasis (VL). Additional work was then conducted to delineate its essential structural features, aiming to improve solubility and safety without compromising activity against VL. While the 4-nitroimidazole portion was specifically required, several modifications to the aryloxy side chain were well tolerated e.g., exchange of the linking oxygen for nitrogen (or piperazine), biaryl extension, and replacement of phenyl rings by pyridine. Several less lipophilic analogues displayed improved aqueous solubility, particularly at low pH, although stability towards liver microsomes was highly variable. Upon evaluation in a mouse model of acute Leishmania donovani infection, one phenylpyridine derivative (37) stood out, providing efficacy surpassing that of the original preclinical lead.

2 Environment ACS Paragon Plus

Page 3 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

INTRODUCTION Neglected tropical diseases (NTDs) affect in excess of one billion people, predominantly in the most impoverished areas of the world, causing more than half a million deaths each year, as well as much physical suffering and social stigma.1,2 Of the 17 major NTDs, the kinetoplastid parasite diseases visceral leishmaniasis (VL), Chagas disease, and human African trypanosomiasis (HAT) are considered amongst the most challenging, due to their high mortality rates and limited treatment options, with little economic incentive for this to change.2 VL is caused by Leishmania donovani (L. don) and Leishmania infantum (L. inf), transmitted by female sand flies; these parasites first multiply inside host macrophages before spreading to and destroying other tissues (e.g., spleen, liver, and bone marrow).3,4 Annual incidence of VL is now estimated at 300000, with infection occurring mainly in rural areas of East Africa, India, Bangladesh and Brazil. Most of the ~35000 deaths each year pass unrecognised, and in East Africa, lethal epidemics take place frequently.5 However, the current treatments for VL are unsatisfactory.3,4 Pentavalent antimonials have been the standard first-line remedy since the 1940s, but these are cardiotoxic and require painful parenteral administration over 1 month; drug resistance has now severely limited their utility in India. A less toxic liposomal form of the antifungal agent, amphotericin B (1; see Figure 1), has proven widely effective but is limited by high cost. The cheaper aminoglycoside antibiotic, paromomycin (2), has variable efficacy, both within and between regions. Finally, the alkyl phospholipid miltefosine (3), a long-acting anticancer drug repurposed for VL, is the sole oral agent; nevertheless, this also suffers from several disadvantages, including cost, teratogenicity, and treatment failures. Hence, there remains an urgent need for more effective, safe, and affordable oral treatments for VL. In the last decade of tuberculosis (TB) and NTD drug discovery, there has been major interest in exploiting potential development shortcuts, via the repurposing of approved drugs, rescue of “failed” clinical agents, or repositioning of

3 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

early or late stage drug candidates.6-8 Indeed, for organizations such as the Global Alliance for TB Drug Development (TB Alliance) and the Drugs for Neglected Diseases initiative (DNDi), phenotypic screening is still viewed as a highly useful, cost-effective strategy to identify “low hanging fruit”.9-11 Recent cooperation between these two organizations in this regard actually provided the foundation for this current study.12 In early collaborative work with TB Alliance (2005-6) we investigated13 several nitroheterobicyclic ring analogues of the bioreductive TB agent pretomanid (PA-824, 4), now in phase III clinical evaluation.14 However, our efforts in the known15 6-nitroimidazooxazole class were curbed following disclosure of the recently approved16 MDR-TB drug delamanid (OPC-67683, 5),17,18 turning our focus for a backup toward the efficacious biaryl analogues of 4.19-21 Both 4 and 5 are activated by an intracellular deazaflavin dependent nitroreductase (Ddn) in Mycobacterium tuberculosis (M. tb), resulting in inhibition of cell wall synthesis as well as respiratory poisoning (via nitric oxide release).22 This mode of action is distinct from the postulated nucleophilic nitro reduction (e.g., by thiolate) of other new nitroaromatic TB agents, such as PBTZ169 (6) and nitrobenzamides, to form reactive nitroso intermediates, which trigger suicide inhibition of an enzyme (DprE1) involved in cell wall biosynthesis.23,24 In late 2009, follow-up interest from DNDi in some promising antileishmanial screening data for several of our earlier nitroimidazole derivatives (above) then led to our nomination of three unoptimised racemic hits, including two from the most potent 6-nitroimidazooxazole class (7 and 8), for proof-of-concept in vivo assessment in a mouse model of VL. In parallel, taking into account the preliminary SAR findings, a further 72 compounds were screened against L. don in a macrophage-based luciferase assay at the Central Drug Research Institute (CDRI, India),25 and 10 of these were later tested in the same mouse model. However, the outstanding in vivo efficacy of 7 was not surpassed by any of the other candidates, and following our synthesis of its enantiomers and ensuing head-to-head assessments of these, the

4 Environment ACS Paragon Plus

Page 4 of 72

Page 5 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

R enantiomer 9 (DNDI-VL-2098)18 was chosen for preclinical evaluation as a potential firstin-class drug candidate for VL.25,26 In this report, we initially detail results relevant to the selection of 9. We then describe the findings of more recent work aimed at discovering backups to 9 having an improved physicochemical/ pharmacological profile and better safety, providing wider SAR conclusions for the nitroimidazooxazole class against TB and three kinetoplastid NTDs, together with a first in vivo appraisal of the best new leads for VL. CHEMISTRY Several earlier investigators15,27 prepared 6-nitro-2,3-dihydroimidazo[2,1-b][1,3]oxazoles (“oxazoles”) by condensing epoxides with the explosive intermediate 2,4-dinitroimidazole. In the interests of improving safety, we first sought alternative conditions using 2-bromo-4nitroimidazole (75; see Scheme 1A/1B).13 Satisfactory results for both 2-H and 2-methyl glycidyl ethers were obtained by neat reaction with 75 in the presence of a suitable base (DIPEA, 105-108 °C) and the resulting (uncyclised) alcohol intermediates were readily transformed into the desired oxazoles 7 and 10-12 in high yield (85-90%) upon treatment with sodium hydride (DMF, 0 °C). Minor quantities (1-3%) of 5-nitroimidazooxazoles were also generated in the first step, with two examples (15 and 87) being isolated for comparison. The requisite epoxides (74, 78, 82,28 and 86) were acquired via standard methods, starting with (4-OCF3) benzylations of glycidol (73) or 2-methylprop-2-en-1-ol (77), or alkylations of 4-(trifluoromethoxy)phenol with epibromohydrin (81) or 3-chloro-2-methylprop-1-ene (84). To synthesise the enantiomers of 12 and 15 (13, 14, 16, and 17), we employed cesium fluoride promoted reactions29 between 4-(trifluoromethoxy)phenol and the chiral glycidyl nosylates 89 and 93 (Scheme 1C). The ether products (90 and 94) were then alternatively coupled with the more accessible 2-chloro-4-nitroimidazole (91), where the use of toluene as cosolvent (DIPEA, 80 °C) enabled better isolated yields of the more thermally labile 5-nitro

5 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

compounds 16 and 17 (15-16%). In the final step, sodium hydride induced cyclisation of the intermediate alcohols (92 and 95) was completed by warming to 20 °C, giving both oxazoles 13 and 14 in high yield (86-87%) and excellent ee (100%). Reaction of the known18,30,31 chiral epoxides 96 and 97 with 4-(trifluoromethoxy)phenol or 4'-fluorobiphenyl-4-ol in the presence of sodium hydride (Scheme 2A) also provided a direct route to the enantiomers of early VL leads 7 and 8, as recorded for 9.18 These epoxides were conveniently obtained by the reaction of 2-chloro-4-nitroimidazole (91) with (R) or (S)-2methylglycidyl 4-nitrobenzenesulfonate (derived from the Sharpless epoxidation of 2methylprop-2-en-1-ol, 77),31 although a multistep approach18 based on the 4-nitrobenzoate ester derivative of (S)-2-methylglycidol reportedly achieved a slightly better ee for 96 (98.8% vs. 95.9%).30 While isolated yields of the final products 9, 18, 44, and 45 were moderate (3344%), these important entities were rapidly produced in excellent chiral purities (>98% ee), and a superior kilogram scale process synthesis of 9 has recently been developed.32 For more efficient generation of additional analogues of 7, we then turned to the known33 racemic epoxide 98 (Scheme 2B). Heating this epoxide with various phenols and sodium hydride furnished the oxazole products 19-23 directly, albeit in variable yields (12-45%), depending on the conditions. Alternatively, reaction of 98 with 4-(trifluoromethoxy)aniline (or its N-methyl derivative) in the presence of anhydrous cobaltous chloride34 gave uncyclised β-anilino alcohols (99 and 100), which could be transformed into the novel oxazoles 24 and 25 by careful treatment with sodium hydride (1.3-1.4 equiv). Next, in order to access larger quantities of the known30 alcohol 26 for the preparation of heteroaryl ethers, we devised an improved plan, comprising TIPS monoprotection of known30 diol 102, ring closure, and desilylation under acidic conditions (Scheme 2C). Diol 102 was obtained from 2-chloro-4-nitroimidazole (91) in two steps (85%), via dihydroxylation of the new alkene 101, and elaborated to the oxazole 104 as planned (92% from 102; two steps).

6 Environment ACS Paragon Plus

Page 6 of 72

Page 7 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

However, attempted removal of the TIPS group using a reported acidic method (1% HCl in 95% EtOH)35 required prolonged heating at 45 °C (4.5 days) and was not clean, leading to a suboptimal yield of alcohol 26 (53%). Fortunately, treatment of 104 with hydrofluoric acid in acetonitrile36 gave a much better result (95% yield), and 26 was easily converted into various heteroaryl ether targets (27, 29-31) via sodium hydride-catalysed SNAr reactions on the haloheterocycles 105-108.37 The isomeric pyridine 28 was also obtained in moderate yield (25%) by reaction of 6-(trifluoromethyl)pyridin-3-ol and sodium hydride with the known30 epoxide 109 (Scheme 2D). By way of comparison, epoxides 98 and 109 each afforded comparable yields (55% and 51%) for scale-up of racemic VL lead 7. In the 2H oxazole series, the synthesis of biphenyl analogues (32-34) was achieved in a straightforward manner (Scheme 3A), starting from epibromohydrin (81), and using methodology described above (except that DMF was employed in the final step to improve solubility). We then envisaged the assembly of related phenylpyridine side chains via Suzuki couplings on bromopyridinyl ether derivatives (e.g., 119 in Scheme 3B). To this end, the known27 alcohol 117 was prepared in three steps from TBS-protected glycidol (113), via condensation with 91, ring closure of alcohol 115, and acid-catalysed desilylation (Scheme 3B). Unfortunately, reaction of 117 with 5-bromo-2-fluoropyridine (118) under the optimised conditions developed for the 2-methyl congener 144 (Scheme 4B) gave a poor yield of ether 119 (9%), and attempted Suzuki couplings on this substrate (even with weak bases) resulted only in decomposition. In an attempt to obtain at least one or two examples featuring this isomer pattern, we next considered reversing the order of these final two steps. Here, the 5aryl-2-fluoropyridines 120 and 121 were quantitatively obtained from bromide 118 but the final alkylations involving alcohol 117 gave very poor yields (30 µM, most >64 µM), in broad agreement with assessments on VERO cells (data not shown).50 In the design of new analogues of 7 and 8, two strategies were employed to enhance aqueous solubility while minimising the risk of major potency loss.20,21 First, we focused on reducing lipophilicity (probed using CLogP predictions derived from ACD LogP/LogD software, version 12.0; Advanced Chemistry Development Inc., Toronto, Canada) via the replacement of benzene rings by pyridine, or closely related heterocycles (viz. pyrimidine, pyrazine, and pyridazine). This approach enabled CLogP reductions of up to 2.3 units (typically ~1.1 units for a single pyridine substitution, although the predicted data displayed some variation between isomers). Second, we evaluated the removal of full aromaticity from the side chain by the incorporation of a saturated heterocycle (e.g., piperazine or piperidine) in combination with a terminal aromatic ring. Arylated cyclic amines are known bioisosteres for biaryl moieties21 and can provide improved solubility via salt formation, as well as by the disruption of molecular planarity and symmetry.54 Kinetic aqueous solubility values at pH=7 were determined for selected active compounds, using dry powder forms, while analogous

11 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

data at pH=1 were measured for examples containing basic functionalities that could form hydrochloride salts (based on calculated pKa estimates derived from similar ACD software). 1. Initial studies relevant to preclinical candidate selection The investigations began in late 2009 with the proof-of-concept in vivo assessment of oxazoles 7 and 8 in a mouse model of VL at the London School of Hygiene and Tropical Medicine (LSHTM). The selection of these two candidates was based on a combination of their promising activities in an L. don-infected macrophage assay conducted at the Swiss Tropical Institute (later confirmed at CDRI) and reasonable conjecture for their better metabolic stabilities in comparison to other similarly active analogues (e.g., 11). In the event, both 7 and 8 provided outstanding efficacy in this L. don model at 50 mg/kg (99.9 to 100% reduction of the liver parasite burden, after oral dosing once daily for 5 days) and subsequent dose-response experiments (Table 3) established that 7 was the most effective lead (ED50s were 3.0 and 4.8 mg/kg for 7 and 8, respectively). This result was consistent with 7 having superior mouse microsomal stability (79% vs. 57% parent remaining after 30 min, Table 3), better aqueous solubility (16-fold), and a more favourable mouse pharmacokinetic profile (lower clearance, greater exposure, and higher oral bioavailability; Table 4 and Supporting Information, Figure S1) in comparison to 8. The more stringent chronic infection hamster model of VL is considered the gold standard in vivo assay due to its similar pathology to human disease.55 Here, 5 and 10 day assessments of 7 and 8 at CDRI led to opposing conclusions on which oxazole was better (Supporting Information, Table S3); therefore, the enantiomers of both compounds were prepared. In the mouse model, the R enantiomer of 7 (9) was greatly superior to the S enantiomer (18) and was slightly better than 7 itself (83%, 8%, and 64% inhibition at 3.13 mg/kg, respectively; Table 3). This was in good agreement with the microsomal stability results (Table 3) and L. inf IC50 data (Table 1) showing a 4.5-fold potency difference between 9 and 18. In contrast,

12 Environment ACS Paragon Plus

Page 12 of 72

Page 13 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

although the R enantiomer of 8 (44) was also better than the S enantiomer (45) in this model, it was less active than 8 itself (36% vs. 68% at 6.25 mg/kg; Table 3), indicating reduced oral exposure (44 and 45 gave sparing aqueous solubilities of ~0.07 µg/mL). A comparative appraisal of 7, 9, and 18 in the CDRI hamster model (Supporting Information, Table S3) confirmed the preeminent activity of the R enantiomer (9) over the S form (18) and 7 itself, at all dose levels.25 Furthermore, dose-response evaluations of 9 and the TB drug delamanid (5) in a similar L. inf infected hamster model at LMPH also corroborated the excellent in vivo efficacy of 9 (>99% inhibition in 3 target organs at 25 mg/kg; Table 5), whereas 5 displayed a weak effect (22-66% at 50 mg/kg), consistent with its poor in vitro potency (IC50 7.1 µM). In view of these findings, the decision was made to proceed with 9 as a lead candidate for VL. While the initial in vivo work was in progress, the oxazole analogues in hand (7, 8, 10-12, 46, and 47) were screened against L. don in the macrophage-based luciferase assay at CDRI.25 Based on these results (Table 1) and some encouraging microsomal stability data (Table 3), two additional compounds were identified as worthy of investigation in the VL mouse model. These were the (6-fold) more potent des-methyl analogue of 7 (12: IC50 0.005 µM) and the 4-trifluoromethoxy analogue of 8 (47: IC50 0.05 µM). However, counterintuitively to our previous work,19 47 was markedly inferior to 8 for VL (only 42% inhibition at 25 mg/kg), while the more potent lead 12 also displayed slightly lower efficacy than 7 (89% inhibition at 6.25 mg/kg; Table 3). To confirm the latter result, we examined the enantiomers of 12 (13 and 14), but surprisingly, while the R form (13) was preferred in vitro (Table 1), both stereoisomers had an equivalent effect in vivo (~75% at 6.25 mg/kg; Table 3). Examination of the pharmacokinetic properties of selected compounds afforded further insight (Table 4), revealing that 12 had a higher rate of clearance than 7, gave a lower exposure level, and had a concomitantly reduced oral bioavailability (52% vs. 79%), likely due in part to its 4-fold lower aqueous solubility (Table 3). Furthermore, while 47 did show

13 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

better metabolic stability than 8 (decreased clearance and longer half-life; Table 4), it also suffered from poor oral absorption (Cmax 0.29 µg/mL; see Supporting Information, Figure S1), resulting in modest oral bioavailability (18% cf. 63% for 8). Again, this was consistent with the 11-fold lower aqueous solubility of 47 compared to 8 (Table 3) and suggests the need for stronger oral formulations when evaluating such sparingly soluble compounds. Thus, the early selection of 9 for advanced development12 was validated by the discovery of inferior in vivo results for the few alternative candidates that were assessed at the time. Following the demonstration of suitable in vivo efficacy and in vivo pharmacokinetics, drug discovery for leishmaniasis typically requires several additional ADME and safety studies.56 For example, many nitroheterocyclic compounds provide positive results in the Ames test for mutagenicity; indeed, the development of an early 6-nitroimidazooxazole lead for TB (CGI17341) was halted following such a discovery.15 Extensive investigation by Otsuka Pharmaceutical Co. established that the introduction of heteroatoms and larger side chain functionality (e.g., 5) overcame this effect18,57 and both 8 and 926 were Ames negative. Other safety factors (e.g., hERG, CYP inhibition, potential for drug-drug interactions) were also broadly acceptable for 9,26 which additionally satisfied such criteria as high stability, low cost of synthesis and suitability for once-per-day oral administration noted in the proposed target product profile for a VL drug.58 Oral treatment for VL offers many important advantages, including convenience and improved patient compliance; it also enables direct delivery of the absorbed drug to the liver, a major parasite population site.26 Therefore, the repositioning of 9 for VL may represent a significant medical breakthrough for this highly neglected disease. 2. SAR of nitroimidazooxazoles for VL In an attempt to better understand the SAR for VL (with a view to further improving druglike features, e.g., solubility and safety), additional oxazoles were prepared and screened, first at CDRI, and then at LMPH. The latter testing confirmed (Table 1) that this class had no

14 Environment ACS Paragon Plus

Page 14 of 72

Page 15 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

obvious potential for treating HAT (T. brucei IC50s >1 µM, most >64 µM), nor any clear utility for Chagas disease (T. cruzi IC50s mostly 1-3 µM, and no new leads were better than 5 or 8), so the major focus on VL was justified. Nevertheless, in light of the recently reported efficacy of fexinidazole (a 5-nitroimidazole derivative) in a mouse model of VL,59 one fundamental question was whether 5- or 6-nitroimidazooxazoles were preferred. However, an inspection of the data for 15-17 (5-nitro isomers of 12-14 above) immediately showed that the 5-nitro analogues were more than two orders of magnitude less effective against L. inf. Having established this, compounds 19-23 initially examined the influence of the phenyl ring substituent. Interestingly, while the single determination CDRI data (L. don) appeared to indicate a clear advantage for 4-trifluoromethoxy (7), this was not supported by the mean IC50 results vs. L. inf, which pinpointed only the 4-phenoxy analogue (23) as being significantly less active, with the remaining compounds (including 7) falling within a ca. 2fold potency range. Replacement of the side chain linking oxygen atom in 7 by nitrogen (24 and 25) was also very well tolerated (L. inf IC50s ≤1.5-fold apart), and, in the case of 24, this allowed a 6-fold solubility improvement (albeit, this compound seemed to be less stable than 7). In contrast, removal of the aryl moiety (alcohol 26; ∆CLogP -3.4 units over 7) led to a 55fold loss in VL activity, suggesting that an aryl-based side chain was necessary (as for TB). Therefore, we next investigated a small set of trifluoromethyl-substituted heteroaryl ethers (27-31). This design concept was based on some partial success with employing trifluoromethylpyridine as a more soluble substitute for trifluoromethoxyphenyl in our previous TB studies.20,21,37 Pleasingly, pyridine 27 (L. inf IC50 0.24 µM) was slightly more potent than both 7 and the direct phenyl equivalent 22, and was 3-fold more soluble than 7 (11 µg/mL). However, less lipophilic analogues of 27 (28-31) gave 2- to 6-fold poorer IC50 values against the parasite, despite in two cases (28, 30) providing superior solubility.

15 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The SAR focus then switched to targets with phenylpyridine side chains, which could be regarded20 as more soluble mimics of the biphenyl lead 8. For completeness, the relatively insoluble 2H biphenyls 32-34 were first evaluated and found to be 2- to 6-fold more active than their 2-methyl counterparts (8, 46, and 47), similar to the 2-fold higher L. inf potency of 12 vs. 7. The 2H phenyl-2-pyridines 35 and 36 had comparable or up to 6-fold reduced activities compared to the biphenyls 33 and 34, respectively. Conversely, 2H phenyl-3pyridine derivatives 37, 40 and 41 were noteworthy for the exceptional potency of the 4fluoro compound 37 (L. inf IC50 0.03 µM), which was 11-fold better than 7 and 25-fold superior to the 4-trifluoromethoxy analogue 41. Similar SAR trends were revealed in the 2-methyl series where, for phenyl-3-pyridine derivatives (51-53), the 4-fluoro compound 51 (L. inf IC50 0.13 µM) was overwhelmingly more active than congeners 52 and 53 (by 40- to 85-fold). By contrast, in the less soluble and more lipophilic (∆CLogP +0.6 units) isomeric phenyl-2-pyridine series (48-50), the 4-fluoro compound 48 (L. inf IC50 2.9 µM) was no better than biphenyl lead 8 or close analogue 49, although the 4-trifluoromethoxy derivative 50 was inferior. It was interesting to observe that neither 37 nor 51 showed enhanced potencies against the other parasites (or TB) and that both compounds displayed ~5-fold better aqueous solubility than 8 at neutral pH, in harmony with their decreased lipophilicities (∆CLogP -1.1 to -1.5 units). This solubility differential over 8 was considerably larger at low pH (>2000-fold; Table 3), suggesting further promise. Therefore the enantiomers of 37 (38 and 39) were also prepared and assessed, together with their 5-nitro isomers (42 and 43). Here, unexpectedly, the S enantiomer 39 was the most active (2-fold over 38), while the 5-nitro analogues were 3 orders of magnitude less effective. Overall, given the superior in vivo efficacy of 8 compared to 47 and the much higher in vitro potencies of 37 and 51, it did appear that a 4-fluoro substituent in the terminal ring was preferred for VL. Therefore, in the remaining compounds (54-59) we retained this element

16 Environment ACS Paragon Plus

Page 16 of 72

Page 17 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

and probed further modifications expected to improve solubility. Commencing with the bipyridines 54-56, the most synthetically challenging isomer 54 was found to be the best (L. inf IC50 0.14 µM), having >20-fold higher potencies than the related phenylpyridine 48 and the biphenyl congener 8, and 13- to 16-fold greater aqueous solubility than these compounds at neutral pH (in accordance with its reduced lipophilicity, ∆CLogP -1.1 to -1.6 units). Importantly, more structurally diverse phenylpiperazine and phenylpiperidine analogues (57 and 58) also retained excellent activities (L. inf IC50s 0.20 and 0.23 µM). As noted above, phenylpiperazine and phenylpiperidine are soluble bioisosteres for biphenyl and these moieties are now appearing in many new drug candidates.60,61 Here, 57 and 58 were respectively 61-fold and 11-fold more soluble than 8 at neutral pH and formed nicely soluble hydrochloride salts at low pH (19-23 mg/mL), in line with their calculated pKa values of 5.9 and 7.3. Finally, based on previous success with an arylpiperazine carbamate analogue of 4 for TB,41 carbamate 59 was studied, but the additional carboxy group proved detrimental for VL (8.5-fold potency loss). There was also a poor tolerance for longer side chains (60 and 5). With these SAR results in hand, the next stage of this work was to assess microsomal stability and in vivo efficacy for the best new leads. Six racemic compounds (pyridine 27, phenylpyridines 37 and 51, bipyridines 54 and 56, and phenylpiperazine 57) were selected as representative of both potency and solubility optimisations for the new side chain classes investigated here. Unsurprisingly, these diverse compounds demonstrated a wide range of metabolic stabilities (Table 3) following 1 h exposures to human and mouse liver microsomes (HLM and MLM). While all compounds showed broadly acceptable stabilities towards HLM (~50% or more parent remaining), phenylpyridine 51 and phenylpiperazine 57 displayed low stabilities toward MLM (14-21% parent), with the others having moderate MLM stabilities (47-69% parent). The most stable compounds were pyridine 27 and bipyridine 56, and a 2-H substituent (37) appeared to reduce the rate of MLM metabolism (cf. 2-methyl derivative 51).

17 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nevertheless, upon in vivo evaluation in the mouse model of VL (dosing orally at 25 mg/kg for 5 consecutive days), all candidates except the least potent one (bipyridine 56) provided essentially total parasite clearance (>99.5%), similar to 7, 8, and 12. At 6.25 mg/kg, the more rapidly metabolised phenylpiperazine 57 demonstrated poor efficacy (16% inhibition), whereas pyridine 27 (83%) was very similar to the des-methyl phenyl ether 12 (89%), bipyridine 54 was highly effective (96%), and phenylpyridines 37 and 51 both achieved a 100% cure (Figure 2 and Table 3). From further dose reductions on 37, 51, and 54, it was subsequently established that the highly potent 2H phenylpyridine 37 was the most active lead (98% inhibition at 1.56 mg/kg, cf. 64

14

43

11

A

Me

OCF3

0.08

0.28

0.42

60

60

0.24

34

12

B

H

OCF3

0.005

0.15

1.5

>64

>64

2.0

73

13

Bd

H

OCF3

0.02

0.33

4.7

>64

>64

0.32

70

14

Be

H

OCF3

0.05

0.50

2.6

>64

>64

3.1

>128

15

C

H

OCF3

48

9.8

33

31

>128

>128

16

Cd

H

OCF3

>64

>64

>64

>64

56

61

17

Ce

H

OCF3

>64

>64

>64

>64

>128

>128

7f

B

Me

OCF3

0.03

0.33

1.8

>64

>64

0.077

55

9g

Bd

Me

OCF3

0.03

0.17

2.6

53

>64

0.046

5.9

18

Be

Me

OCF3

0.03

0.77

1.6

38

>64

2.5

27

19g

B

Me

H

0.14

0.30

1.8

>64

>64

0.16

22

20

B

Me

F

0.12

0.17

1.8

>64

>64

0.11

15

21

B

Me

Cl

0.12

0.20

2.4

57

>64

0.038

41

22

B

Me

CF3

0.28

0.41

2.5

>64

>64

0.058

46

56 Environment ACS Paragon Plus

Page 57 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

23

B

Me

OPh

0.25

2.2

4.5

8.9

>64

0.11

48

24

D

H

OCF3

0.06

0.32

7.1

50

>64

0.36

24

25

D

Me

OCF3

0.08

0.50

0.90

12

>64

0.40

9.7

26f

E

18

52

>64

>64

16

116

27

B

Me

2

CF3

0.21

0.24

1.7

>64

>64

0.038

28

28

B

Me

3

CF3

0.30

0.55

2.9

>64

>64

0.48

45

29

B

Me

2,3

CF3

1.4

2.2

>64

>64

0.91

106

30

B

Me

2,5

CF3

0.43

1.4

>64

>64

0.14

70

31

B

Me

2,6

CF3

0.50

1.5

>64

>64

0.47

>128

32

F

H

F

0.02

1.8

2.8

>64

>64

0.025

>128

33

F

H

CF3

0.11

3.8

2.8

>64

>64

0.040

>128

34

F

H

OCF3

0.01

0.83

1.4

3.1

>64

0.039

>128

35

F

H

2’

CF3

2.4

19

2.2

45

58

57

36

F

H

2’

OCF3

2.1

5.0

10

5.2

47

5.3

48

37

F

H

3’

F

0.07

0.03

1.3

>64

>64

0.049

42

38

Fd

H

3’

F

0.06

7.1

>64

>64

0.037

2.7

39

Fe

H

3’

F

0.03

0.95

>64

>64

0.46

3.8

40

F

H

3’

CF3

0.01

1.5

0.94

>64

>64

0.024

>128

41

F

H

3’

OCF3

0.01

0.76

1.4

4.2

>64

0.037

>128

42

Gd

H

3’

F

48

34

>64

54

>128

>128

43

Ge

H

3’

F

49

22

>64

55

84

>128

8

F

Me

F

0.06

3.3

0.52

9.2

>64

0.043

11

44

Fd

Me

F

0.06

2.3

2.8

>64

>64

0.046

4.0

45

Fe

Me

F

0.35

4.2

0.58

>64

>64

0.080

>128

46

F

Me

CF3

0.06

11

1.5

1.0

>64

0.081

>128

47

F

Me

OCF3

0.05

5.3

0.89

0.85

>64

0.088

64

48

F

Me

2’

F

0.20

2.9

0.65

>64

>64

0.024

35

49

F

Me

2’

CF3

0.11

3.6

0.85

16

>64

0.018

4.5

50

F

Me

2’

OCF3

0.07

26

2.4

8.9

>64

0.018

19

57 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 58 of 72

51

F

Me

3’

F

0.04

0.13

0.99

>64

>64

0.036

35

52

F

Me

3’

CF3

0.06

5.2

2.5

>64

>64

0.023

>128

53

F

Me

3’

OCF3

0.03

11

1.6

3.9

>64

0.025

>128

54

F

Me

2’,2

F

0.15

0.14

1.2

40

>64

0.053

24

55

F

Me

2’,3

F

0.34

0.31

5.6

52

>64

0.091

14

56

F

Me

3’,3

F

0.09

0.57

7.9

>64

>64

0.37

39

57

H

0.16

0.20

1.1

>64

>64

0.13

41

58

I

0.63

0.23

1.1

12

>64

0.29

60

59

J

0.81

1.7

3.6

>64

>64

0.23

29

60f

K

13

0.40

3.4

>64

0.051

26

5g

Kd

7.1

0.43

1.7

>64

0.033

2.3

a

IC50 values for inhibition of the growth of Leishmania donovani, Leishmania infantum, Trypanosoma cruzi, and Trypanosoma brucei, or for cytotoxicity toward human lung fibroblasts (MRC-5 cells). bEach value (except the single test L. don data) is the mean of at least two independent determinations. For complete results (mean ± SD) please refer to Supporting Information. cMinimum inhibitory concentration against M. tb, determined under aerobic (MABA)50 or hypoxic (LORA)51 conditions. dR enantiomer. eS enantiomer. fRef 30. g Ref 18.

58 Environment ACS Paragon Plus

Page 59 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Table 2. In vitro antiparasitic and antitubercular activities of ring A analogues of 7

IC50a,b (µM) compd

Fm

MICc,b (µM)

Y

L. inf

T. cruzi

T. bruc

MRC-5

MABA

LORA

7

A

6-NO2

0.33

1.8

>64

>64

0.077

55

61

A

6-H

>64

2.1

>64

>64

81

>128

62

A

6-Br

>64

22

33

34

55

51

63

A

6-CHO

>64

33

>64

>64

>128

>128

64

A

6-CH2OH

>64

>64

>64

>64

>128

>128

65

A

6-SMe

57

9.4

>64

38

54

66

66

A

6-SO2Me

57

>64

>64

>64

>128

>128

67

A

5-Br

30

0.13

33

26

61

61

68

A

5-SMe

>64

0.58

33

25

27

70

69

A

5-SO2Me

>64

20

>64

>64

>128

>128

70

B

NO2

>64

0.59

20

>64

>128

>128

71

B

Br

>64

13

60

43

46

>128

72

C

NO2

>64

4.7

34

>64

52

>128

a

IC50 values for inhibition of the growth of Leishmania infantum, Trypanosoma cruzi, and Trypanosoma brucei, or for cytotoxicity toward human lung fibroblasts (MRC-5 cells). bEach value is the mean of at least two independent determinations. For complete results (mean ± SD) please refer to Supporting Information. cMinimum inhibitory concentration against M. tb, determined under aerobic (MABA)50 or hypoxic (LORA)51 conditions.

59 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 60 of 72

Table 3. Aqueous solubility, microsomal stability, and in vivo antileishmanial efficacy data for selected analogues

aq solubilitya

microsomal stabilityb

in vivo efficacy against L. don (mouse)

(µg/mL)

[% remaining at 1 (0.5) h]

(% inhibition at dose in mg/kg)c

compd pH=7 pH=1

H

M

Ham

25

12

0.91

(99)

(76)

(54)

>99

13

0.96

74

14

1.2

76

7

3.6

(86)

(79)

(49)

9

2.4

(92)

(89)

18

2.6

(84)

(70)

24

23

11d

27

11

11d

84

69

>99

83

28

50

30

27

31

4.2

32

0.059

37

1.3

526

60

47

>99

100

38

1.1

260

57

54

52

45

36

39

1.4

299

61

41

59

46

13

8

0.23

(88)

(57)

6.25

3.13 1.56 0.78

44

0.075

36

45

0.068

6

47

0.021

48

0.19

2.7

51

1.1

54

3.1

99

64

31

7

18 (54)

>99

>99

83

49

11

(64)

84

19

8

(79)

>99

0.39

89

>99

(42)

100

12.5

93

>99

98

68

(91)

(100)

42

634

58

21

>99

100

93

34

46

71

52

>99

96

94

47

60 Environment ACS Paragon Plus

Page 61 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

55

3.3

19

56

10

57

14

19100

58

2.6

23200

5

0.31

116

82

62

91

49

14

>99

16

55

a

Solubility (µg/mL) in water (pH = 7) or 0.1 M HCl (pH = 1) at 20 °C, determined by HPLC (see Experimental Section). bPooled human (H), CD-1 mouse (M), or hamster (Ham) liver microsomes; data in brackets are the percentage parent compound remaining following a 30 min incubation. cDosing was orally, once daily for 5 consecutive days; data are the mean percentage reduction of parasite burden in the liver. dUnstable under assay conditions.

61 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 62 of 72

Table 4. Pharmacokinetic parameters for selected compounds in male Swiss Albino mice

oral (12.5 or 25 mg/kg)a

intravenous (1 mg/kg)

compd

CL

Vdss

t1/2

AUClast

(µg/mL) (mL/kg/ (L/kg) min)

(h)

(µg⋅h/mL)

C0

Cmax

Tmax

AUClast

Fb

(µg/mL) (h) (µg⋅h/mL) (%)

12

0.61

13

2.2

3.2

1.14

1.8

1.0

14.8

52

7

0.88

9.5

1.7

2.2

1.69

4.1

4.0

33.5

79

9

0.36

11

2.6

2.7

1.39

1.3

2.0

15.8

91

8

0.30

28

7.5

3.8

0.48

0.70

6.0

7.58

63

47

0.34

7.6

5.7

11

2.13

0.29

6.0

9.81

18

a

The oral dose for 9 was 12.5 mg/kg; the remainder were dosed at 25 mg/kg. bOral bioavailability, determined using dose normalised AUClast values.

62 Environment ACS Paragon Plus

Page 63 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Table 5. In vivo efficacy of 5 and 9 against L. infantum in the early curative hamster model (LMPH)

dosage a compd (mg/kg)

liver

spleen

bone marrow

3

20

81.8

92.0

84.7

5

50

65.7

51.3

21.7

25

53.9

42.9

37.3

12.5

32.5

24.8

23.7

50

100

100

100

25

100

99.9

99.7

12.5

99.0

98.7

94.0

9

a

% inhibition in target organs

All compounds were dosed orally once daily for 5 consecutive days.

63 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Structures of various antileishmanial or antitubercular agents

64 Environment ACS Paragon Plus

Page 64 of 72

Page 65 of 72

Figure 2. Comparative in vivo efficacy in the mouse VL model: a) 6.25 mg/kg; b) 1.56 mg/kg a)

b) 120

120

100

100

80

80

% Inhibition

% Inhibition

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

60 40

60 40 20

20

0

0 12 13 14

7

9 18 27 37

8 44 45 51 54 57

Test compound

8 7

10 9

58 37

72 51

75 54

Test compound

65 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

Figure 3. Comparison of 4, 9, 18, and 44 with rifampicin in the acute TB infection mouse model

1.0E+07 1.0E+06

CFU/mouse

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1.0E+05 1.0E+04 1.0E+03 1.0E+02 1.0E+01 T3

T10

T31

4

9

18

44

RMP

Test compound

66 Environment ACS Paragon Plus

Page 66 of 72

Page 67 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Scheme 1a

a

Reagents and conditions: (i) 4-OCF3BnBr, NaH, DMF, 0-20 °C, 7-21 h; (ii) DIPEA, 105108 °C, 6.5-15 h; (iii) NaH, DMF, 0 °C, 0.7-1.4 h or 0-20 °C, 2-3 h; (iv) m-CPBA, Na2HPO4, CH2Cl2, 0-20 °C, 3-3.5 h; (v) 4-OCF3PhOH, K2CO3, acetone, 59 °C, 36-41 h; (vi) 4OCF3PhOH, CsF, DMF, 20 °C, 35-45 h; (vii) DIPEA, toluene, 79-80 °C, 24 h.

67 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Scheme 2a

a

Reagents and conditions: (i) ArOH, NaH, DMF, 0-55 °C, 2-3 h, or 0-80 °C, 5-30 min; (ii) 4OCF3PhNH2 or 4-OCF3PhNHMe, CoCl2, CH3CN, 63-65 °C, 24-52 h; (iii) NaH, DMF, 0-20 °C, 1.3-2.8 h (or 0 °C, 2 h); (iv) K2CO3, DMF, 65 °C, 33 h; (v) OsO4, NMO, CH2Cl2, 20 °C, 8 h; (vi) TIPSCl, imidazole, DMF, 20 °C, 6.5 d; (vii) 40% HF, CH3CN, 20 °C, 11 h; (viii) ArX (105, 106, 107, or 108), NaH, DMF, 0-20 °C, 1-2.5 h.

68 Environment ACS Paragon Plus

Page 68 of 72

Page 69 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Scheme 3a A O

O

i

O2N

O 81

N

+

110 (89%)

Br

NH

O2N

Cl

iii

N N

N O 112 (81%)

111 (83%)

I

O

N

O2N

O

Cl

91

I

I

OH

ii

iv

compounds 32-34 of Table 1 (77-84%)

B O

O2N

ii

OH

+ 91

OTBS

+

N

113

N

N

O OTBS 114 (2.5%)

iii

N

O2N

Cl

N

115 (71%)

N

O

OR

116: R=TBS (72%) 117: R=H (98%) vi

v

Reagents for final step: Br

N

F

R

O

N

120: R=CF3 (100%); 121: R=OCF3 (100%)

Br

O

N

O2N 118

compounds 35 and 36 of Table 1 (0.8-0.9%)

N

iv

F

vi

N

O2N

OTBS

119 (9%)

C N

N

vii

HO

O

Br

N

iv Br

O 123 (79%)

O

Br

N

126: R=F 130: R=CF3 134: R=OCF3 (85-86%)

81

O S 89 O O

+

91

N

R

O

O S 93 O O

F + 91

F O 138 (87%)

xi

N

O x

(75-85%)

127: R=F; 131: R=CF3; 135: R=OCF3 (68-74%)

136 (89%)

NO2

O

Cl

O

x

compounds 37, 40 and 41 of Table 1

iii

N

O2N

O NO2

R

125: R=F; 129: R=CF3 133: R=OCF3 (95-98%) R

OH N

ii

D O

HO

124: R=F; 128: R=CF3 132: R=OCF3 (97-99%)

O ix

R

O

122

O

N

viii

O

+ 91 N F

X Y

N

O2N N

iii N

Cl

O

137: X=H, Y=OH (67%) 139: X=OH, Y=H (68%) + compounds 42 and 43 of Table 1 (10-12%)

compounds 38 and 39 of Table 1 (87-89%)

xi

a

Reagents and conditions: (i) 4-IPhOH, K2CO3, acetone, 59 °C, 52 h; (ii) DIPEA, 100-105 °C, 6-13 h; (iii) NaH, DMF, 0-20 °C, 1.7-4 h; (iv) ArB(OH)2, (DMF), toluene, EtOH, 2 M Na2CO3, Pd(dppf)Cl2 under N2, 85-90 °C, 0.8-4.5 h; (v) 1% HCl in 95% EtOH, 20 °C, 6.5 h; (vi) ArF (118, 120, or 121), NaH, DMF, 0-20 °C, 2.5-3.7 h; (vii) EtOCH2Cl, K2CO3, DMF, 0-20 °C, 24 h; (viii) 1.25 M HCl in MeOH, 54-55 °C, 5 h; (ix) ArOH (125, 129, or 133), K2CO3, MEK, 65 °C, 39-49 h; (x) 125, NaH, DMF, 20 °C, 40 h; (xi) DIPEA, toluene, 80 °C, 24 h.

69 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 70 of 72

Scheme 4a A

OH

I i

O

O

+

Cl

O

140

ii O2N

NH

O2N

N 75

141 (73%)

I iii

N N

Br

O

Br

N

N

Br

26

I

compounds 8, 46 and 47 of Table 1 (84-90%) N

O2N

O

N

iv

143 (92%)

142 (85%)

v

OH + F

O O

N

B O2N

N

O2N

O

O N 144 (46%)

118

vi

N

iv

Br

compound 54 of Table 1 (36%)

O F

B N

O

145

compounds 48-50 and 55 of Table 1 (69-80%) C OH vii

N

O2N

O

N

N

N

O2N N

Br

+

N

O2N

O O

N

O

Br

98

Br

iv N

147 (45%)

146 (10%)

Br

compounds 51-53 and 56 of Table 1 (53-75%)

D HN

N

O2N

O

N

148: Z=N 150: Z=CH

109 E O2N N

ix OH O

26

F

viii

compound 59 of Table 1 (86%)

OH

N

O2N N

Cl

N

Z

+

iii Cl

N Z

149: Z=N 151: Z=CH

compounds 57 and 58 of Table 1 (75-79%)

F

(54-88%)

a

Reagents and conditions: (i) 4-IPhOH, K2CO3, NaI, DMF, 72 °C, 24 h; (ii) DIPEA, 108 °C, 14.5 h; (iii) NaH, DMF, 0 °C, 1-1.5 h; (iv) ArB(OH)2, (DMF), toluene, EtOH, 2 M Na2CO3 (or 2 M KHCO3), Pd(dppf)Cl2 under N2, 78-92 °C, 0.75-3 h; (v) NaH, DMF, 0-20 °C, 2.5 h; (vi) 145, Cs2CO3, CuCl, dppf, DMF, Pd(dppf)Cl2 under N2, 85 °C, 4.3 h; (vii) 6-Br-pyridin3-ol, NaH, DMF, 0-50 °C, 4 h; (viii) MEK, (CHCl3), 65-80 °C, 18-72 h; (ix) triphosgene, Et3N, THF, 0-20 °C, 1.7 h, then 148, THF, 20 °C, 3.5 h.

70 Environment ACS Paragon Plus

Page 71 of 72

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Scheme 5a A

OCF3

O

OH NH

+ O

N

86

i

OCF3

N

NO2

N

NO2

152

B 86

Br

+

Br

N

N

Br

O

N N

H

iii

O

7 D

OCF3 + 86

86

+

O2N

N NH 158

NO2

O

O

N

O

N

O2N

N

O

OH N N

O

OCF3 +

NO2

O 159 (6%)

Br

+

compound 70 of Table 2 (34%)

N N N

OCF3

O2N

O O

compound 69 of Table 2 (86%)

OCF3

N N

i

Br

O2N

OCF3

compound 66 of Table 2 (66%)

MeO2S v

compound 68 of Table 2 (44%)

i

O

compound 65 of Table 2 (33%)

OCF3

N

E

v

N

MeS N

N

N 157

O

compound 67 of Table 2 (89%)

O

Br

MeS vi

N NH

O2N

ii

iii

compound 62 of Table 2 (87%)

OCF3

156 (29%)

O

OCF3

O

N

O

N

OCF3 compound 63 of Table 2 (39%) iv compound 64 of Table 2 (87%) C N O O2N O N

OH

155 (69%)

N

Br

O

O

compound 61 of Table 2 (83%)

OCF3

N

+ Br ii

154

O

Br

OCF3

N

ii

compound 61 of Table 2 (43%)

153 (44%)

OH i

NH

O

N

+

O

153

N

N N

OCF3

O O

compound 71 of Table 2 (21%)

OCF3

O O

compound 72 of Table 2 (66%)

a

OCF3

Reagents and conditions: (i) DIPEA, (toluene), 104-109 °C, 12-16 h; (ii) NaH, DMF, 0-20 °C, 80 min (for 61), or 55 °C, 4 h (for 62), or 45 °C, 2.5 h (for 67); (iii) nBuLi, THF, -78 °C, 80 min, then DMF or (MeS)2, -78 to 20 °C, 4 h (for 63), or -78 to 0 °C, 18 h (for 65), then aq citric acid; (iv) NaBH4, EtOH, 0 °C, 1 h; (v) m-CPBA, Na2HPO4, CH2Cl2, 0-20 °C, 18-26 h; (vi) MeSH, Et3N, MeOH, 0-10 °C, 3 h.

71 Environment ACS Paragon Plus

Journal of Medicinal Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of Contents graphic

72 Environment ACS Paragon Plus

Page 72 of 72